| | news

"All models are wrong, but some are useful"

Adam Riess dans son bureau à l'Université John Hopkins. | Gail Burton / AP Keystone

Nobel Prize winner, Adam Riess is one of the pioneers in the standard model of cosmology. In an interview with Christian Speicher, he explains why he is no longer fully convinced of this model.

Mr Riess, not many cosmologists get the privilege of contributing to a path-breaking cosmological perspective. You played a major role in this end of the 1990s. What happened then?

At that time, it was believed that the universe was essentially made up of matter, whose gravitational pull slowed the expansion of the universe possibly to such an extent that the universe would one-day collapse again. To our astonishment, however, we noticed that the expansion was not slowing down, but accelerating. This finding let us conclude that matter is not the most important ingredient of the universe. Instead, there must be something that counteracts the attraction of matter and accelerates expansion. We refer to this something as dark energy.

What was your first impression when you saw the data?

I thought they were wrong and we made a mistake in the measurement. What we saw was mind-boggling. But as a scientist, you have a duty to publish your results when you have done your best and excluded all sources of error. Only in this way can other cosmologists try to confirm or refute these results.

How long did it take the cosmologists to accept the new cosmological view of the universe?

It took about five years. This was mainly due to observations of the cosmic microwave background radiation with the WMAP satellite of NASA, followed by the Planck satellite of ESA. These confirmed that there really must be more matter and energy in the universe than the five percent we can see. These measurements fit very well to the image of the accelerated expansion. In this respect, two lines of evidence have come together here. The fact that another group besides ours had observed evidence of an accelerated expansion of the universe also certainly contributed to the acceptance. So our discovery could not be based on a simple measurement error.

The idea that we live in a universe that is expanding faster and faster under the influence of dark energy is now regarded as the standard model of cosmology. Can you be satisfied with this model? After all, 95 percent of it is based on forms of matter and energy that we cannot specify in more detail.

The standard model is basically a very simple model. We see two completely unrelated gravitational phenomena in the universe. On the one hand, we observe strong cohesion between galaxies and clusters of galaxies, stronger than the gravity of ordinary matter would suggest. This tells us that there must be another form of matter besides ordinary matter – the dark matter. On the other, we see that things are moving away from each other on larger scales. Matter does not do such a thing by itself, not even dark matter. That leads us to the dark energy. We do not need more than these two components. Although the standard model is very simple, it is very successful. But that does not mean it is complete.

Do you see parallels to particle physics? It also has a very successful standard model that leaves fundamental questions unanswered.

There is a well-known saying. All models are wrong, but some are useful. A model rarely provides a complete description of reality. Usually, there is always something missing. The decisive factor is whether the model allows verifiable computations and predictions. In this sense, the standard models of both particle physics and cosmology are useful.

While the standard model of particle physics has been as solid as a rock for decades, doubts about the standard model of cosmology are growing. What is the reason for that?

One reason for this is that the current expansion rate of the universe seems to be greater than that suggested by the standard model of cosmology. By the precise measurement of the cosmic background radiation we believe to know in which state the universe was shortly after the Big Bang. With the standard model we can extrapolate how fast the universe should expand today, more than 13 billion years later. However, measurements of the Hubble constant - a measure of the current expansion rate - provide a greater value. Consequently, the model that links the past to the present seems to be wrong, of course, provided that the measurements are correct.

Well, are they correct?

I would say the measurements are quite reliable now. We have been working on more accurate measurement of the Hubble constant for many years. The measured values have hardly changed. But the error bars have become smaller and smaller. On the other side, measurements of cosmic background radiation have also become more accurate. Therefore, the measurement errors today are several times smaller than the discrepancy between the extrapolated and the actually measured expansion rate of today's universe. That must make us think.

That sounded quite different a few years ago. Even though this discrepancy was apparent then, you hesitated to question the standard model. What made you change your mind?

In the meantime, it has become possible to make cross-comparisons with other experiments. This resulted in a consistent picture. The measurements of the early and the present conditions do not fit together if one assumes that the universe was formed and developed in such a way as the standard model of cosmology suggests. This can hardly be traced back to measurement errors, unless several independent effects would have been plotted in unison.

What conclusions must be drawn from this? Is it time to bid farewell to the standard model of cosmology?

I would not go that far. It could rather amount to refining the standard model. Like I said, it is a pretty simple model. For instance, it is assumed that dark matter consists of particles that do not interact, collide and destroy each other. One of those assumptions may not be true. It could just as well be that the dark energy is not a cosmological constant, but becomes stronger with time. In this case, the standard model would predict a greater expansion rate for the universe today. There are other ideas, as well. However, all of the proposals are clarifications of the standard model, and not an entirely new model.

Other researchers continue there. They are considering modifying the theory of gravity.

Most cosmologists today assume that our gravitational theory is correct. So far, we do not know a single case, where the general theory of relativity has failed. Nevertheless, we must be open to the possibility of this paradigm being wrong. This would have a direct impact on the dark domain of the standard model. I do not consider this the first option. In addition, one must reckon that it is not so easy to develop alternative theories of gravity. It is not enough to explain the strong cohesion of galaxies and the accelerated expansion of the universe. The theory must also be able to explain the orbits of the planets around the sun and much more.

So far, one has searched in vain for particles of dark matter. Are you confident they will be found one day?

I am not sure, but I am optimistic. In any case, it is worthwhile to continue searching for these particles. I find it quite legitimate to consider other possibilities. Nevertheless, they must pass the many tests that the dark matter has already passed.

The Hubble constant measurements call for a modification of the standard model. As you mentioned, the dark energy may become stronger over time. What consequences would that have?

None in the immediate future. However, the consequences would be drastic over long periods of time. This is because no object that is held together by gravity would be stable in the long run. We call this the "Big Rip". First clusters of galaxies would be torn apart, followed by galaxies and finally, planets like the earth. That would be a very different ending than the "Big Crunch" – the inevitable big collapse we expect in a universe dominated by matter.

Sortir de la crise, la newsletter qui aborde les enjeux de la sortie de crise selon une thématique différente

Lire aussi